$C^*$-algebra Net: A New Approach Generalizing Neural Network Parameters to $C^*$-algebra
This introduces a new theoretical framework for neural networks that could enhance model adaptability and efficiency, but it appears incremental as it builds on existing mathematical concepts.
The paper tackles the problem of generalizing neural network parameters to $C^*$-algebra-valued ones, enabling continuous combination of models and efficient feature learning, and shows application to density estimation and few-shot learning with limited samples.
We propose a new framework that generalizes the parameters of neural network models to $C^*$-algebra-valued ones. $C^*$-algebra is a generalization of the space of complex numbers. A typical example is the space of continuous functions on a compact space. This generalization enables us to combine multiple models continuously and use tools for functions such as regression and integration. Consequently, we can learn features of data efficiently and adapt the models to problems continuously. We apply our framework to practical problems such as density estimation and few-shot learning and show that our framework enables us to learn features of data even with a limited number of samples. Our new framework highlights the potential possibility of applying the theory of $C^*$-algebra to general neural network models.